Inhibition of Angiotensin-Converting Enzyme Activity by Flavonoids:Structure-Activity Relationship StudiesLigia Guerrero1,2,Julia´n Castillo3,Mar Quin˜ones1,Santiago Garcia-Vallve´1,4,Lluis Arola1,4,Gerard Pujadas1,4,Begon˜a Muguerza1,4*1Department of Biochemistry and Biotechnology,Rovira i Virgili University,Tarragona,Spain,2Department of Research,Nutrition and Innovation,ALPINA S.A,Bogota´, Colombia,3Department of Research and Development,Nutrafur S.A.,Murcia,Spain,4Centre Tecnolo`gic de Nutricio´i Salut(CTNS),TECNIO,CEICS,Avinguda Universitat, Reus,Catalonia,SpainAbstractPrevious studies have demonstrated that certain flavonoids can have an inhibitory effect on angiotensin-converting enzyme (ACE)activity,which plays a key role in the regulation of arterial blood pressure.In the present study,17flavonoids belonging to five structural subtypes were evaluated in vitro for their ability to inhibit ACE in order to establish the structural basis of their bioactivity.The ACE inhibitory(ACEI)activity of these17flavonoids was determined by fluorimetric method at two concentrations(500m M and100m M).Their inhibitory potencies ranged from17to95%at500m M and from 0to57%at100m M.In both cases,the highest ACEI activity was obtained for luteolin.Following the determination of ACEI activity,the flavonoids with higher ACEI activity(i.e.,ACEI.60%at500m M)were selected for further IC50determination.The IC50values for luteolin,quercetin,rutin,kaempferol,rhoifolin and apigenin K were23,43,64,178,183and196m M, respectively.Our results suggest that flavonoids are an excellent source of functional antihypertensive products.Furthermore,our structure-activity relationship studies show that the combination of sub-structures on the flavonoid skeleton that increase ACEI activity is made up of the following elements:(a)the catechol group in the B-ring,(b)the double bond between C2and C3at the C-ring,and(c)the cetone group in C4at the C-ring.Protein-ligand docking studies are used to understand the molecular basis for these results.Citation:Guerrero L,Castillo J,Quin˜ones M,Garcia-Vallve´S,Arola L,et al.(2012)Inhibition of Angiotensin-Converting Enzyme Activity by Flavonoids:Structure-Activity Relationship Studies.PLoS ONE7(11):e49493.doi:10.1371/journal.pone.0049493Editor:Jian R.Lu,The University of Manchester,United KingdomReceived July19,2012;Accepted October9,2012;Published November21,2012Copyright:ß2012Guerrero et al.This is an open-access article distributed under the terms of the Creative Commons Attribution License,which permits unrestricted use,distribution,and reproduction in any medium,provided the original author and source are credited.Funding:This study was supported by grants from the Ministerio de Educacio´n y Ciencia of the Spanish Government(AGL2008-00387/ALI)and from the Universiti Rovira i Virgili-Banco Santander(2011LINE-12).The funders had no role in study design,data collection and analysis,decision to publish,or preparation of the manuscript.Competing Interests:The authors have declared that no competing interests exist.*E-mail:begona.muguerza@urv.catIntroductionCardiovascular disease(CVD)is the most important cause of death among the industrialized societies[1].Hypertension,which is estimated to affect one-third of the Western population,is one of the major risk factors for CVD[2].In spite of its significance, hypertension remains poorly controlled[3],and approximately two-thirds of hypertension is undetected or inadequately treated [4].Lifestyle modifications,including changes in dietary habits, have substantial effects on risk factors for CVD,such as hypertension[5].The renin–angiotensin–aldosterone system is a key factor in the maintenance of arterial blood pressure.One of its main components is the angiotensin-converting enzyme(ACE)[EC 3.4.15.1][6],which is a glycosylated zinc dipeptidyl-carboxypep-tidase whose main function is to regulate arterial blood pressure and electrolyte balance through the renin–angiotensin–aldoste-rone system[7].There are two isoforms of ACE that are transcribed from the same gene in a tissue-specific manner.In somatic tissues,ACE exists as a glycoprotein composed of a single large polypeptide chain of1,277amino acids,whereas in sperm cells,it occurs as a lower-molecular-mass glycoform of701amino acids.The somatic form consists of two homologous domains(the N and C domains),each of which contains an active site with a conserved HEXXH zinc-binding motif[8],where the Zn2+is bound to the two motif histidines as well as to a glutamate24 residue downstream the last motif histidine[9].The testis ACE (tACE)is identical to the C-terminal half of somatic ACE,with the exception of a unique36-residue sequence that constitutes its amino terminus[10].The two domains differ in their substrate specificities,inhibitor and chloride activation profiles,and physiological functions[11].Thus,mice expressing only the N domain of ACE show a low blood pressure phenotype that is very similar to ACE knockout mice[12],and ACE inhibition with an N-domain-specific inhibitor(i.e.,RXP407)has no effect on blood pressure regulation[13].On the other hand,mice that are homozygous for a mutation that inactivates the somatic ACE N domain,but not the C domain,retained a phenotype that was indistinguishable from that of wild-type mice with regards to blood pressure and renal function[14].Therefore,the inhibition of the C domain appears to be necessary and sufficient for the control of blood pressure and cardiovascular function,which suggests that the C domain is the dominant angiotensin-converting site.As an exopeptidase,ACE catalyzes the conversion of angiotensin I into the potent vasoconstrictor angiotensin II[15].In addition,ACE catalyzes the inactivation of the vasodilator bradykinin[16]. Therefore,the inhibition of this enzyme can generate anantihypertensive effect.In fact,synthetic ACE inhibitors,such as captopril and enalapril,are widely used for the treatment of cardiovascular and renal disease,for the secondary prevention of coronary artery disease,and for the treatment of heart failure[17]. However,side effects such as cough,angioneurotic edema and deleterious effects in pregnancy have been associated with the clinical use of ACE inhibitors[18,19].Therefore,the investigation of new,natural product-based ACE inhibitors could greatly benefit hypertensive patients.A number of extracts and compounds obtained from plants have been identified as in vitro ACE inhibitors[20,21].These beneficial effects have largely been ascribed to the presence of flavonoid molecules,which generation of chelate complexes within the active center of ACE[22].Flavonoids are polyphenol molecules of low molecular weight;the basic structure is a2-phenyl benzopyrone in which the three-carbon bridge between the phenyl groups is usually cyclized oxygen[23,24].Flavonoids can be differentiated into several subfamilies according to their degree of unsaturation and the degree of oxidation of the oxygenated heterocycle and can be characterized as flavanones,flavones, flavonols,isoflavones,flavanols(essentially,flavan-3-ols)and anthocyanidins,all of which are the most relevant for the human diet[23,25].Different studies have revealed the important role that flavonoid structure plays in its biological function;the position and number of substituents in the flavonoid basic structure significantly affects the antiproliferative,cytotoxic,antioxidant, and anti-enzymatic activities of such molecules[26–28]. Previous studies have shown that certain flavonoids exhibit a capacity to inhibit different zinc metalloproteinases[29,30], including ACE.In fact,micromolar concentrations of different flavonoids,such as anthocyanins[31,32],flavones[33],flavonols [33–35],and flavanols[36],have been shown to inhibit50%of ACE activity.Furthermore,the ACE-inhibitory(ACEI)activity of different foods and plant extracts rich in flavonoids has also been demonstrated by in vitro[37,38],studies and by in vivo studies in hypertensive rats[39,40]and humans[41].The preliminary structure-activity relationships(SAR)studied in some flavonoids (flavanols and flavonols)generally attribute the observed effect either to the distribution of free hydroxyl groups[33,35,42]or in the number of monomers units forming the corresponding procyanidins[36].However,the key molecular flavonoid sub-structures that dictate effective ACE inhibition activity have not yet been characterized.The objective of this work was to define the key flavonoid structural elements that are required for ACE inhibition activity through the determination of the ability of17flavonoids belonging to five structural subtypes(i.e.,5flavanones,2flavan-3-ols,1 isoflavone,6flavones and3flavonols;including potassium salts for 1flavanone and for1flavone)to inhibit ACE.To achieve this goal,the in vitro ACE inhibition activity of these17flavonoids was measured,and the corresponding results were used to establish SAR for these molecules.Afterwards,protein-ligand docking studies were used to describe the molecular basis for most significant SAR results.Materials and MethodsChemicalso-aminobenzoylglycyl-p-nitro-phenylalanylproline(o-ABz-Gly-Phe(NO2)-Pro)was purchased from Bachem Feinchemikalien (Bubendorf,Switzerland).Five units of Angiotensin-I Converting Enzyme from rabbit lung and ZnCl2were obtained from Sigma (Barcelona,Spain).All flavonoids(assay.90%purity)used in this study were kindly provided by Nutrafur S.A.(Murcia,Spain), except for the catechin,luteolin and genistein,which came from Sigma-Aldrich Quı´mica(Barcelona,Spain).Flavonoids were solubilized in dimethyl sulfoxide(DMSO)and prepared daily.InTable1.HPLC analysis of the flavonoid samples used in the current study.Main content of the minorflavonoidsCompound content(%)1(%)2Other minor flavonoids Naringenin94.7Naringin(0.6)Naringenin-7-glucoside Naringenin K93.8Naringin(0.9)Naringenin-7-glucoside Naringin95.2Narirutin(1.1)Poncirin,Naringenin Apigenin97.2Rhoifolin(0.7)Apigenin-7-glucoside Apigenin K96.5Rhoifolin(0.8)Apigenin-7-glucoside Rhoifolin96.1Naringin(0.4)ApigeninGenistein97.2Daidzein(1.1)GenistinLuteolin95.9Eriodictyol(0.8)Luteolin-7-glucoside Hesperetin94.7Hesperidin(1.2)Hesperetin-7-glucoside Diosmetin93.8Diosmin(2)HesperidinDiosmin95.2Hesperidin(1.8)HesperetinCatechin98.1Other catechins Epicatechin98.5Other catechins Quercetin95.7Rutin(1.5)IsoquercitrinRutin97.1Isoquercitrin(1.1)QuercetinKaempferol96.8Quercetin(0.8)Kaempferol-3-glucoside1Absolute value as is.2The reference%assay is referred to the absolute content as is of the main flavonoid.doi:10.1371/journal.pone.0049493.t001doi:10.1371/journal.pone.0049493.g001all experiments,the final concentration of DMSO was 0.4%.Distilled water was obtained from a Millipore Milli-Q H system.Preparation of SolutionsThe 0.45mM buffered substrate solution (o -Abz-Gly-p-Phe(NO 2)-Pro)and 150mM Tris-acid buffer solution containing 1.125M NaCl (pH 8.3)were prepared daily.Flavonoid solutions (100and 500m M)were prepared in DMSO (0.4%)daily.The 0.1U/mL ACE solution stock was prepared in glycerol:water (1:1),aliquoted and stored at 220u C.The 0.1m M ZnCl 2stock solution was prepared and stored at 4u C.The ACE working solution was prepared daily by diluting it in 150mM Tris buffer (pH 8.3)containing 0.1m M ZnCl 2.Chromatographic Analysis and Quantification of Flavonoid CompoundsFor the elucidation and quantification of the main flavonoids present in each sample,we modified a previously published method [27].All the samples were dissolved in DMSO in the ratio of 5mg/mL,and the resulting solutions were filtered through a 0.45m m nylon membrane.The HPLC equipment was a Hewlett-Packard Series HP 1100equipped with a diode array detector.The stationary phase was a C18LiChrospher 100analytical column (25064mm i.d.)with a particle size of 5m m (Merck,Darmstadt,Germany)thermostated at 30u C.The flow rate was 1mL/min and the absorbance changes were simultaneously monitored at 280and 340nm.The mobile phases for chromato-graphic analysis were:(A)acetic acid:water (2.5:97.5)and (B)acetonitrile.A linear gradient was run from 95%(A)and 5%(B)to 75%(A)and 25%(B)for 20min;changed to 50%(A)and (B)for 20min (40min,total time);changed to 20%(A)and 80%(B)for 10min (50min,total time),and finally re-equilibrated for 10min (60min,total time)to the initial composition.Table 1resumes the global HPLC profile of the different samples used in this study.Measurement of ACE-inhibitory ActivityACEI activity was measured by a fluorimetric assay following the method of Sentandreu and Toldra ´[43]with some modifica-tions [44].A volume of 160m L of 0.45mM buffered substrate solution in 150mM Tris-acid buffer containing 1.125M NaCl,(pH 8.3)was mixed with 40m L of the flavonoid solution (with 0.4%DMSO for the blank samples)and 40m L of ACE solution (0.1U/mL),and the mixture was incubated at 37u C.Fluorescence was measured after 30min in 96-well microplates (Thermo Scientific,Rochester,NY)using a multiscan microplate fluorim-eter (Biotek.FL 6800).Microplates (Thermo Scientific,Rochester,NY)were used in this assay.The excitation and emission wavelengths were 360and 430nm,respectively.The activity of each sample was tested in technical and biological triplicate.The ACEI activity was calculated using the following formula:%Inhibitory activity ~100|A {C ðÞ=A {B ðÞ,where A is the fluorescence without the flavonoid solution,B is the fluorescence without ACE and C is the fluorescence in the presence of both ACE and the flavonoid solution.A flavonoid solution of 500m M was selected on the basis of previous studies [33,35],and the concentration of 100m M was chosen because it is within the physiological concentration range.An ACEI activity higher than 60%at 500m M concentration was used as a selection criterion for the IC 50(the flavonoid concentration required to inhibit the original ACE activity by 50%).The IC 50of each selected flavonoid was tested in technical and biological triplicate.The results from three experiments are expressed as the mean 6SD and were performed in different platelet samples.Molecular Modeling StudiesFigure 2.(A)Effect of different flavonoids on Angiotensin Converting Enzyme (ACE)activity.Purified lung ACE was preincubated at 37u C for 30min in the presence of 100m M of flavonoids or DMSO as a control.The results are expressed as the percentage of ACE inhibition.The plot represents the mean result 6SD from three experiments.(B )Effect of different flavonoids on Angiotensin Converting Enzyme (ACE)activity.Purified lung ACE was preincubated at 37u C for 30min in the presence of 500m M of flavonoids or DMSO as control.The results are expressed as the percentage of ACE inhibition.The plot represents the mean 6SD from three experiments.doi:10.1371/journal.pone.0049493.g002Table 2.IC 50values obtained for the selected flavonoids.IC 50value 1(m M)Apigenin K196Rhoifolin (apigenin 7-O-glycoside)183Kaempferol178Rutin (quercetin 3-O-glycoside)64Quercetin 43Luteolin231The IC 50value represents the concentration of each compound that inhibits ACE activity by 50%.doi:10.1371/journal.pone.0049493.t002Figure 3.Structural diagram which quantitatively assesses the effect of the addition or elimination of different structural elements from the flavonoid core on the ACEI activity of luteolin at 100m M.According to these data,the significance order was:double bond C2=C3(absence:291%activity by comparing naringenin vs.apigenin).49-O-methoxylation (presence:278%by comparing diosmetin vs.luteolin)<4-carbonyl group (absence:274%by comparing epicatechin vs.luteolin).39-hydroxylation (absence:257%by comparing apigenin vs.luteolin).3-hydroxylation (presence:244%by comparing quercetin vs.luteolin).3-O-glycosylation (presence:-36%by comparing rutin vs.luteolin).doi:10.1371/journal.pone.0049493.g003Flavonoid structures were either obtained from ChemSpider (/)or drawn with Marvin Sketch v5.9.0(ChemAxon Kft.,Budapest,Hungary;http://www. /).All flavonoid structures were further set up with LigPrep v2.5(Schro¨dinger LLC,Portland,USA;http:// )following three steps:(1)using Epik software[45]to generate all possible protonation states within a pH range of4.067.0and selecting the‘‘add metal binding states’’Table3.Intermolecular interactions between ACE inhibitors and the tACE binding site.LISI-NOPRIL ENAL-APRILATCAPTO-PRILRXP-A380SELENO-CAPTOPRIL KAF KAW lisW-S FII-ALUTE-OLINQUER-CETIN RUTINKAEMP-FEROLS29Gln28133NE2NE2NE2NE23H HThr2823333His353NE2NE2NE2NE2NE2NE2NE2NE2NE23333Glu37633Asp4533Lys511NZ NZ NZ NZ NZ NZ NZ NZ NZHis513NE2NE2NE2NE2NE2NE2NE2NE2NE2Tyr520OH OH OH OH OH OH OH OH OHS29/S19Ser284Val3793333Val380333333S19Glu1623OE2 Asn2773Asn374Asp377OD1S1Glu143Val351Ala354O O33333O333O OSer35533333333Trp357Lys368Glu384OE2OE2OE2OE23OE2OE2OE2OE2333Phe5123333333Ser516Tyr523OH OH3OH3OH OH OH OH OH OH OH3S1/S2Val5183333333S2Phe3913333Glu403Arg522Other Thr1663 Trp2793Ala356N N N NHis383333NE23NE2NE2NE2NE23His387NE233NE233NE2NE2His410333Glu411333OE133OE1Asp41533OD23OD133Lys454NZ NZPhe457333333333Phe527333Data used for lisinopril,enalaprilat,captopril,RXPA380,selenocaptopril,KAF,KAW,lisW-S and FII-A was obtained from the LigPlot+diagrams that are available at the PDBsum resource(/pdbsum/)for PDB files1O86,1UZE,1UZF,2OC2,2YDM,3BKK,3BKL,3L3N and2XY9,respectively.Data for luteolin,quercetin, rutin and kaempferol was obtained by applying LigPlot+to the structure of their predicted complexes with tACE.Hydrophobic contacts are indicated by a check mark whereas hydrogen bonds are indicated with the label of the protein atom that is involved.doi:10.1371/journal.pone.0049493.t003option to generate possible ligand-metal binding states among metalloproteins;(2)generating tautomers at the previously given pH range;and (3)determining chiralities from the ligand’s 3D structure.All the protein-ligand docking studies performed in this work were performed with Glide v5.7(Schro ¨dinger LLC.,Portland,USA;)[46,47]with extra precision (i.e.,with GlideXP;[48]).Before docking the flavonoids,the shape and properties of the ACE binding site were represented by several different sets of fields on a grid.This grid was made of a box that has default dimensions around the location of the experimental pose of the inhibitor (i.e.,RXPA380),and the inhibitor forms a complex with tACE at PDB file 2OC2[49].No constraints were set while building the grid.Default settings for the rest of the grid set-up options were used.During the protein-ligand docking,a maximum number of 5poses per ligand were obtained.Then,the most reliable pose was selected (irrespective of its glide score)by taking advantage of the information provided by (1)the other experimental complexes between ACE and ACE inhibitors available in the PDB (Protein Cate Base)()and (2)the SAR results obtained in the current study for flavonoids.No constraints were imposed during the docking except for those flavonoids in which all docked poses were far from the area predicted for either luteolin or quercetin.In those cases,the AC ring location of their selected poses was used to restrict their docking (i.e.,luteolin was used during apigenin and diosmetin docking,whereas quercetin was used to restrict the docking of rutin and kaempferol).Restricted docking with rhoifolin failed as a consequence of the steric hindrance between the ACE binding site and the 7-O-glycoside substituent;therefore,no docking results are reported for rhoifolin.The results for predicted poses were compared with known experimental poses for ACE inhibitorsbyFigure 4.Best predicted pose for luteolin (panel A)at the tACE binding site and the relative location to experimental poses for the ACE inhibitors lisinopril (panel B),enalaprilat (panel C),and captopril (panel D).All of the panels in this figure are in the same relative orientation to allow for easier comparisons between the poses.Residues at the ACE binding site are colored according to the subsite where they belong (i.e.,residues from the S29,S29/S19,S19,S1and S1/S2subsites are colored in red,cyan,magenta,green,brown,white and yellow,respectively).Other important residues that have not been classified in any pocket are colored in white.Carbon atoms for the ligands are shown in yellow to make them more easily distinguishable from the binding site residues.Dashed lines are used to show intermolecular hydrogen bonds (in red)or charge-charge interactions (in blue).doi:10.1371/journal.pone.0049493.g004comparing their intermolecular interactions with the ACE binding site.With that aim,LigPlot diagrams for the experimental ACE-inhibitor complexes were obtained from the PDBsum website (/pdbsum/)and compared with equivalent diagrams derived from LigPlot+[50]for predicted ACE-flavonoid complexes.Results and DiscussionSeventeen flavonoids were evaluated for their ACEI activity. The structures of all compounds studied are represented in Figure1.All the flavonoids were studied at concentrations of100 and500m M(see Figures2A and2B,respectively).The maximum inhibitory potencies were57%at100m M and95%at500m M.At both concentrations,the highest ACEI activity was exhibited by luteolin.The relative inhibitory potencies for the most active flavonoids(i.e.,ACEI higher than30%)were luteolin.apigenin K.rutin.rhoifolin.quercetin.kaempferol.apigenin.diosme-tin.narigenin K.epicatechin.genistein.hesperetin and diosmin for500m M;and luteolin.kaempferol.rutin.rhoifolin.querce-tin for100m M.The rest of the flavonoids exhibited ACEI activities lower than30%.The IC50value was obtained for each flavonoid that exhibited an ACEI activity higher than60%at 500m M(i.e.,luteolin,apigenin K,rutin,rhoifolin,quercetin and kaempferol;see Table2).These IC50values were found to be in the23to196m M range(with luteolin being the flavonoid with the highest ability to inhibit ACE activity).In recent years,flavonoids have gained a great amount of interest with regards to their potential for cardiovascular protection.In fact,many epidemiological studies associate an increased consumption of foods and beverages rich in flavonoids with a reduced risk of CVD death[51–53].Additionally,several of these flavonoids or their derivatives(i.e.,diosmin,rutin and quercetin)are widely used as pharmaceutical agents for their vasoprotective properties(i.e.,Daflon500and Venorutom)[54]. Flavonoids are based on the structure of phenyl-benzopyrone and differ from one another in terms of hydroxyl,methoxyl or glycosylated substituents,the position of the benzenoide(B-ring) substituent relative to the C-ring,the degree of unsaturation and the types of sugars that are attached[55].We evaluated the inhibitory effects on the ACE activity of a group of flavonoids from five different structural types(see Figure1).The inhibitory effects of certain flavonoids on ACE activity that have been reported in other studies were confirmed[37].Many of the flavonoids that were tested could inhibit ACE in the micromolar range[22,56]. However,as was expected,significant differences were observed in the ACEI activity depending on the flavonoid structure[25,36]. Although the ACEI activity of these flavonoids does not reach the potency of drugs commonly used in the treatment of hypertension, food products with moderate ACEI activity(i.e.,an ACEI index higher than70%)may be considered as naturally functional foods [57]if it is also taken into account that the regular dietary intake of polyphenols could be as high as1g/day[58,59].Moreover, functional foods containing these natural compounds would not be expected to have the side effects associated with synthetic drugs used in hypertension control[60].Our evaluation of the abilities of different flavonoids to inhibit the activity of ACE confirmed that the principal structural features for their inhibitory activity are as follows:(a)the double bond between C2and C3at the C-ring;(b)the catechol group in the B-ring(39,49-dihydroxy)[61];and(c)the cetone group at the C4 carbon on the C-ring[which is a functional group that has been observed to be essential for inhibiting ACE[62].According to these general considerations,we analyzed and evaluated the SAR derived from our results.Our data confirm that a distinguishing feature for ACE inhibition by flavonoids is the presence of an unsaturated2–3bond conjugated with a4-oxo-function,aside from the39,49-catechol B-ring pattern,as occurs mainly in luteolin,quercetin and rutin.However,it is important to analyze the specific,qualitative and quantitative influence of each one of these three sub-structures in the SAR results.The Significance of the C2=C3Double Bond in the C-ring:Flavone vs.FlavanoneAs previously mentioned,the presence of a C2–C3double bond seems to be essential for the molecule to exert a significant ACEI activity.Two main factors would explain this fact.First,the molecular electronic distribution would allow the maintenance of a definitive structural conjugation,from the B-ring to the A and C rings,in contrast to the flavanone structure,with which this definitive structural conjugation is not possible.Second,the spatial,or the maintenance of a nearly planar structure,would disappear if this bond was saturated, producing a flavonoid skeleton with an obtuse angle,which would be variable depending on the rest of the constituents of the molecule.Our data confirm previous findings that suggest that a nearly planar flavonoid structure is an important factor in the inhibition of ACE[55].In fact,all flavanones included in this study,both aglycones(naringenin and hesperetin)and glycosides(naringin),are not as effective as flavones on ACEI activity.This difference can be observed,more specifically,by comparing the results between the flavone apigenin and its corresponding structurally similar flavanone,naringenin(and, although the difference is on another scale,between diosmetin and hesperetin)where the absence of the C2=C3double bond in naringenin involves a91%reduction of ACEI activity at 100m M vs.apigenin(see Figures2A and3).B-ring Pattern:Catechol Group vs.Monohydroxy Group and O-methylationThe presence of several hydroxyl groups in the flavonoids seems be important for the extent of inhibition of the zinc metallopro-teinases[29].Additionally,the exact position of this group has been revealed to be very important for ACE inhibition. Hydroxylation at the49-position of the B ring seems to be of particular relevance,and in addition,the presence of a catechol group in the B ring(39,49-dihydroxy)appears to be fundamental to achieving an increased ACE inhibitory activity,as occurred in luteolin(as well as quercetin and rutin),which presented the highest ACEI efficiency(see Figure2and Table2).Luteolin has also been described as the most effective flavonoid for inhibiting other metalloproteinases(aminopeptidases),such as MMP-1and MMP-2[30].Consequently,the presence of a catechol at the B-ring should be considered to be very important;indeed,the absence of the39hydroxyl group in apigenin causes a57% reduction of ACEI activity at100m M relative to the luteolin(see Figures2A and3).A similar reduction of activity occurs with the flavonols quercetin and kaempferol,where the absence of catechol in the kaempferol resulted in a4-fold increase in the IC50relative to quercetin(see Table2).Additionally,the characteristic esterification of flavonoids in the 49position significantly reduces ACEI activity,as occurs when the 49-hydroxyl group of luteolin is methylated to generate diosmetin (an esterification that causes a78%reduction of ACEI activity at 100m M;see Figures2A and3).。
